ARTICLE
Principle and pragmatism on the Constitutional Court of South Africa
Theunis Roux*
Though lacking in public support, the Constitutional Court of South Africa (CCSA) today finds itself in a position of relative institutional security. At the same time, it has built up an enviable reputation among constitutional courts in new democracies for the technical quality of its jurisprudence or legitimacy in the legal sense. This essay attempts to explain how this situation has come about by developing a theoretical account of the relationship between legal legitimacy, public support, and institutional security, and then using this account to interpret the CCSA's record from 1995 to 2006. The defining feature of South African politics over this period has been its domination by a single political party. In this context, the theoretical account suggests, the CCS A should largely have been able to ignore its lack of public support in favor of managing its relationship with the political branches. In particular, one would expect the CCSA to have traded off gains in legal legitimacy, achieved by principled decision making, against considerations of the likely impact of its decisions on its institutional security. An examination of some of the CCSA's major decisions reveals that it, indeed, has acted strategically in this way, both in politically controversial cases, where it has used its flexible separation-of-powers doctrine to avoid direct confrontation with the political branches, and in more routine cases, where it has developed a number of context-sensitive review standards.
Introduction
A little more than a decade after deciding its first case,1 the least that can be said about the Constitutional Court of South Africa (CCSA) is that it is still handing down decisions the political branches do not always like.2 By some
'Director, South African Institute for Advanced Constitutional, Public, Human Rights and International Law; extraordinary professor, Centre for Human Rights, University of Pretoria; honorary professor, University of the Witwatersrand. I would like to thank David Bilchitz, Patrick Lenta, Stuart Woolman, Vicki Jackson, and Frank Michelman for comments on earlier drafts of this essay. Email: theunis@saifac.org.za
1 State v Zuma 1995 (2) SALR 642 (CC).
2 See, e.g., Minister of Home Affairs v Fourie; Lesbian andGay Equality Project v Minister of Home Affairs 2006 (1) SALR 524 (CC) (holding the common law definition of marriage and section 3(1) of the Marriage Act 25 of 1961 unconstitutional to the extent that they discriminated against homosexual couples) (hereinafter Fourie); Minister of Health vNew Clicks South Africa (Pty) Ltd. 2006 (2) SALR 311 (CC) (setting aside various regulations relating to the pricing of medicines).
© The Author 2008. Oxford University Press and New York University School of Law. 106 All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org l«C0N, Volume 7, Number 1, 2009, pp. 106-138 doi:10.1093/icon/mon029 Advance Access publication November 17, 2008
Roux   |   Principle and pragmatism on the Constitutional Court of South Africa 107
accounts, the CCSA has been remarkably successful, with a reputation among constitutional courts in new democracies second to none. The truth, of course, lies somewhere in between and depends on whom you talk to and on what the criteria for success are taken to be. Among legal academics, both inside and outside South Africa, the CCSA is generally regarded as having made a credible start on its work of interpreting the two postapartheid constitutions.3 Although social surveys suggest that the Court does not enjoy a great deal of public support,4 this fact may be attributed to the peculiar nature of South African politics, in which a dominant political party frees the CCSA from the need to court public opinion.5 That party, the African National Congress (ANC), has periodically criticized the judiciary,6 but it has not, as yet, threatened to close the CCSA down, despite several significant policy reversals.7 If not universally liked, therefore, the CCSA is today at least relatively secure. A convincing explanation of how this situation has come about promises to contribute both to our understanding of constitutional courts in new democracies and to some of the enduring debates about the legitimacy of judicial review.
5 S. Afr. (Interim) Const. 1993 and S. Afr. Const, 1996. The leading commentaries on South African constitutional law are Constitutional Law of South Africa (2d ed., Stuart Woolman et al. eds., Juta 2004-2008); and Iatn Currie & Johan de Waal, The Bill of Rights Handbook (5th ed., Juta 2005). Though critical in certain respects, both of these commentaries regard the CCSA's record as based on legally plausible interpretations of the postapartheid constitutions. Outside South Africa, assessments of the CCSA's record have been consistently favorable. See, e.g., Cass R. Sunstein, Designing Democracy: What Constitutions Do 221-337 (Oxford Univ. Press 2001) (approving the CCSA's judgment in Government of the Republic of South Africa v. Grootboom 2001 (1) SALR 46 (CC) hereinafter Grootboom); Gregory S. Alexander, The Global Debate over Constitutional Property: Lessons for American Takings Jurisprudence 149-182 (Univ. Chicago Press 2006) (approving the CCSA's constitutional property rights jurisprudence); Mark S. Kende, The South African Constitutional Court's Embrace of Socio-economic Rights: A Comparative Perspective, 6 Chap. L. Rev. 13 7 (2003) (approving the balance struck in the CCSA's socioeconomic rights jurisprudence between rights enforcement and deference to democratic decision making).
4 See James L. Gibson & Gregory A. Caldeira, Defenders of Democracy?: Legitimacy, Popular Acceptance, and the South African Constitutional Court, 65 J. Pol. 1 (2003) (reporting 2 7.9 percent "attentive" public support for the CCSA in 1997, i.e., among citizens who had heard about the Court); James L. Gibson, The Evolving Legitimacy of the South African Constitutional Court, in Justice and Reconciliation in Post-Apartheid South Africa, ch. 9 (Antje du Bois-Pedain & Francois du Bois eds., Cambridge Univ. Press forthcoming 2008) (reporting 34 percent public support for the CCSA in 2004).
5 The ANC won 69.68 percent of the vote in the 2004 election.
5 The best example of this is the ANC National Executive Committee's 93rd anniversary statement of Jan. 8, 2005, in which it accused certain members of the judiciary, without naming them, as out of touch with the aspirations of South Africa's black majority. The statement may be accessed at http://www.anc.org.za/ancdocs/pr/2005/pr0108.html.
7 In addition to Grootboom, the other decision often cited as a significant policy reversal for the ANC is Minister of Health v Treatment Action Campaign (No. 2) 2002 (5) SA 721 (CC) (reviewing and declaring unconstitutional the national Department of Health's program on mother-to-child-transmission of HIV) (hereinafter Treatment Action Campaign).
108 I-CON
January 2009
Vol. 7: 106
Section 1 of this essay assumes that the legitimacy of judicial review (in its legal sense) depends on a court's capacity to decide cases according to forms of reasoning acceptable to the legal community of which it is a part. Legal legitimacy thus understood may be distinguished from two closely related concepts with which it is sometimes confused: public support (confidence in the court among the population as a whole) and institutional security (the court's capacity to resist real or threatened attacks on its independence). Any attempt to compare the record of the CCSA with the record of constitutional courts in other new democracies must take into consideration the uniquely South African configuration of these three factors.
Section 2 argues that political science accounts of constitutional adjudication in new democracies have much to teach legal theorists. The limitation of such accounts, however, is that they lack any real conception of legal legitimacy and, consequently, have little appreciation for the restraining influence of legal doctrine on the behavior of constitutional courts. The problem with currently available theories of judicial review, on the other hand, is that none of them is directed at constitutional courts in new democracies. What is required, therefore, is a new account, drawing on some of the political science insights but expressed in terms acceptable to legal theory.
A fully developed theory of judicial review in new democracies is beyond the scope of this essay. Section 2 nevertheless suggests a way forward by contrasting the particular situation of constitutional courts in new democracies with two well-known theories of judicial review in mature democracies: Ronald Dworkin's conception of courts as "forums of principle" and Richard Posner's claim that appellate courts in the United States should, and often do, act pragmatically. At a theoretical level, this section argues, some combination of principle and pragmatism seems likely to provide the best way for a constitutional court in a new democracy to establish its legal legitimacy while safeguarding its institutional security. "Principle," because deciding cases according to law is what legitimates courts in the legal sense; and "pragmatism," because constitutional courts in new democracies, given the inherent weakness of their position, must perforce temper their commitment to principle with strategic calculations about how their decisions are likely to be received.
The last part of this essay is devoted to showing how the CCSA's record since 1995 can be explained as having been driven by just such a mixture of principle and pragmatism, and that this, in turn, explains how the CCSA has managed to establish its legal legitimacy without undermining its institutional security. Three aspects of its record are considered: (1) cases where the CCSA was able to exploit the political context to hand down decisions of principle in the face of contrary public opinion or determined opposition by the political branches; (2) cases in which this was not possible, and where the CCSA was accordingly forced to compromise on principle to avoid direct conflict with the political branches; and (3) cases in which the CCSA converted conceptual tests that were seemingly required by the constitutional text into more context-sensitive, multifactor balancing tests.
Roux   |   Principle and pragmatism on the Constitutional Court of South Africa 109
1. Clarifying the key concepts: Legal legitimacy, public support, and institutional security
The description of the CCSA's current situation with which this essay began consists of three interlinked claims about the legal legitimacy of its record, the extent of its public support, and its capacity to withstand attacks on its independence. The purpose of this section is to clarify these three claims and to explore their interrelationship in the special circumstances of South Africa. The reason for this exercise is (a) to facilitate cross-country comparison by describing the political and institutional context in which the CCSA has been operating; and (b) to set appropriate limits on the extent to which the findings of this study may be generalized to other constitutional courts.
In the literature on judicial review, "legal legitimacy" refers to the plausibility (rather than correctness) of a judicial decision according to applicable standards of legal reasoning.8 The statement that the CCSA has succeeded in establishing its legal legitimacy thus amounts to a claim that its decisions are generally regarded as having been founded on plausible interpretations of the postapartheid constitutions. I do not attempt to establish an independent basis for this claim but rely, instead, on the overwhelmingly positive assessment of the CCSA's record by local and foreign legal academics.9
Legal legitimacy may be distinguished from sociological legitimacy, which has three distinct meanings in the literature: "institutional legitimacy" (public support for the court despite disagreement with particular decisions), "substantive legitimacy" (public support for particular decisions), and "authoritative legitimacy" (acquiescence in particular decisions, with or without a belief in their correctness).10 Of these three meanings, the first sense of sociological legitimacy is the most relevant to this essay. My concern, as I have said, is to explain the record of a court that is widely admired by legal academics but which has never enjoyed much public support. As with the claim about the CCSA's legal legitimacy, I do not attempt to establish an independent basis for this claim but rely, rather, on social surveys indicating low public support for the CCSA, both at the beginning and also toward the end of the period under review.11
The third key concept, "institutional security," is understood here to mean the CCSA's capacity to survive political attacks on its independence. It is thus akin to "authoritative legitimacy" insofar as it includes the political branches' acquiescence in the CCSA's decisions. However, it is a wider concept than this, since attacks on the CCSA's independence may also include such things as
5 See, e.g., Richard H. Fallon, Jr., Legitimacy and the Constitution, 118 Harv. L. Rev. 1787, 1794-1795 (2005); John C. Yoo, In Defense of the Court's Legitimacy, 68 U.Chi. L. Rev. 775, 782 (2001).
9 See the literature cited supra note 3.
10 See Fallon, supra note 8, at 1828.
11 See Gibson & Caldeira, supra note 4; and Gibson, supra note 4.
110 I-CON
January 2009
Vol. 7: 106
public statements calculated to undermine its reputation, the actual or threatened reduction of its powers, and the appointment of more politically compliant judges. Whereas public support may be assessed by way of social surveys, institutional security is a function of the CCSA's ability to withstand actual or threatened attacks and may be inferred both from the frequency of such attacks and from a qualitative assessment of the CCSA's and other political actors' responses to these attacks.
Thus defined, legal legitimacy, public support, and institutional security are clearly interrelated, albeit not in any simple or predictable way. A court that achieves some measure of legal legitimacy may become more popular and thus more secure, though not necessarily. Conversely, a court may enjoy considerable public support and hence institutional security, without ever having acted legitimately in the legal sense. The precise relationship between legal legitimacy, public support, and institutional security will vary from country to country. It is possible, however, to hypothesize at least two basic rules about the relationship between these factors that should hold for most situations. First, it would seem to be fruitless for a constitutional court in a new democracy to pursue legal legitimacy at the expense of its institutional security, unless the price to be paid for its institutional security was too high. In all but the most extreme cases, in other words, it would seem to make sense for a court to trade off some gains in legal legitimacy in exchange for protecting its institutional security. The extreme case, where this rule would not apply, would be one where the court's failure to decide the case in accordance with its constitutional mandate would forever destroy its reputation for legally legitimate decision making. Such cases have proven to be very rare in the history of strong-form judicial review.12
The second basic rule about the relationship among legal legitimacy, public support, and institutional security is that institutional security typically follows from public support. In most multiparty democracies, a constitutional court that enj oys public support is unlikely to face threats to its position because there would be no political advantage in making or implementing such threats. Likewise, an unpopular constitutional court may become subject to threats by political actors seeking one or another advantage. This rule does not hold for all cases, however. In certain contexts, a constitutional court may be able to afford a substantial loss in public support without becoming institutionally
12 The best example is the Indian Supreme Court's failure to resist the executive's suspension of the writ of habeas corpus during the 1975-1977 emergency. But even in this case, the Indian Supreme Court has survived with its reputation largely intact. See Pratap Bhanu Mehta, The Rise of Judicial Sovereignty, 18 J. Democracy 70, 71, 79 (2007). The same may be said of the U.S. Supreme Court's decision in Bush v. Gore, 531 U.S. 98 (2000). The term "strong-form judicial review" was coined by Mark Tushnet; see, e.g., Mark Tushnet, Weak Courts, Strong Rights: Judicial Review and Social Welfare Rights in Comparative Constitutional Law 18 (Princeton Univ. Press2007). It is used here to refer to the power that certain courts have to strike down laws for unconstitutionality, as opposed to the power merely to declare an incompatibility between an impugned law and the constitution.
Roux   |   Principle and pragmatism on the Constitutional Court of South Africa 111
insecure. This is, in fact, the situation in South Africa, where the ANC mediates the CCSA's lack of public support in a one-party-dominant democracy.13 Because South Africans vote overwhelmingly for the ANC for a variety of reasons related to its history as a former liberation movement, the CCSA's lack of public support is unlikely to be translated into votes for a rival political party.14 In this context, it may be assumed, the CCSA's overriding concern should be to manage it relationship with the political branches.15 Provided that there were some reason why the political branches preferred the CCS A to decide a particular policy issue, the mere fact that the Court's decision was unpopular would not undermine its institutional security. Even decisions that directly thwarted the political branches' policy choices would not necessarily be the subject of political reprisals, so long as the CCSA fulfilled some function useful to the political branches over the long run.16
In an article in this journal, Lynn Berat argued that the CCSA "has failed to acquire institutional legitimacy."17 If this argument was meant to refer to the CCSA's lack of public support, it is correct and clearly supported by the social survey that Berat cites.18 This lack of support should not, however, be equated with a lack of institutional security. The proposed amendment to the CCSA's jurisdiction, which Berat cites as evidence of "the government's hostility to the Constitutional Court,"19 in fact contemplates the expansion of the CCSA's powers.20 While it is possible to construe this proposal as an attempt by the ANC to
13 See The Awkward Embrace: One-Party Domination and Democracy, chs. 1,4, 10, & 11 (Hermann Giliomee & Charles Simkins eds., Tafelberg 1999); Herman Giliomee, South Africa's Emerging Dominant Party Regime, 9 J. Democracy 124(1998).
14 It may also be that constitutional courts in new democracies are relatively immune to low public support because the political elite, even where power changes hands, regards the support of external interests (such as foreign investors) as more important. I am indebted to Patrick Lenta for this point.
15 The Indian case is, by all accounts, exactly the opposite. In that country, the Supreme Court's institutional security flows not from careful management of its relationship with the political branches but from high levels of public support. See Mehta, supra note 12, at 75.
16 Cf. Charles Black, The People and the Court: Judicial Review in a Democracy 56-86 (Macmillan 1960) (describing the "legitimating" function of judicial review during the New Deal era in the United States). For a similar argument in relation to the CCSA, see Theunis Roux, Legitimating Transformation: Political Resource Allocation on the South African Constitutional Court, 10 Democratization 92 (2003) (arguing that the CCSA's legitimacy and the legitimacy of the political branches' transformation efforts are locked into a relationship of mutual dependence).
17 See Lynn Berat, The Constitutional Court of South Africa and Jurisdictional Questions: In the Interest of Justice?, 3 Int'lJ. Const. L. (I-CON) 39, 74 (2005).
18 Id. at 72 (citing Gibson & Caldeira, supra note 4).
19 Berat, supra note 17, at 74.
20 Seethe draft Constitution Fourteenth Amendment Bill, 2005, Bill 22B-05 (GA). An earlier draft of this bill is discussed in Carole Lewis, Reaching the Pinnacle: Principles, Policies and People for a Single Apex Court in South Africa, 21 S. Afr. J. Hum. Rts. 509 (2005).
112 I-CON
January 2009
Vol. 7: 106
take control of the South African judiciary through control of the CCSA, the evidence for this reading has yet to be produced.21 On the contrary, on every occasion on which a member of the executive has criticized the CCSA or threatened to disobey one of its rulings, the ANC has been forced quickly to reaffirm its commitment to judicial independence.22 Far from evidencing a lack of institutional security, such public skirmishes demonstrate a considerable capacity on the part of the CCSA to withstand attacks on its position.
In summary, then, the CCSA today is widely admired for the technical legal quality of its judgments and is relatively secure from political attack, if lacking in public support. The latter fact may be explained by historical and contextual factors. The CCSA's position of relative institutional security, on the other hand, is not so easily explained, especially when considered in combination with its reputation for legally legitimate decision making. How has the CCSA managed to establish this reputation without undermining its institutional security? The next section attempts to lay a theoretical basis for answering this question by comparing a leading political science account of judicial review in new democracies with two well-known legal-theoretical accounts of judicial review in the United States.
2. Theorizing judicial review in new democracies
Over the last twenty years, the political science literature on courts, as encapsulated in the field of judicial politics, has come under the steady influence of rational choice models.23 In terms of this approach,24 courts are viewed neither as neutral interpreters of the law nor as ideologically driven policy makers but, rather, as strategic actors, whose ability to write their policy preferences into law is constrained by the institutional context in which they find themselves.
In one of the most fully developed versions of this approach—by Lee Epstein, Jack Knight, and others—courts are assumed to adjust their decisions according to their expectations about other political actors' policy preferences, and how likely it is that their policy preferences will survive interbranch conflict.25
21 The draft bill was withdrawn after public criticism by former Constitutional Court judges and human rights lawyers sympathetic to the ANC; however, it looks set for reintroduction after a decision to this effect at the ANC's December 2007 party conference in Polokwane.
22 See, e.g., the discussion of the events surrounding the Treatment Action Campaign case in section 3 infra.
23 See Lee Epstein & Jack Knight, Toward a Strategic Revolution in Judicial Politics: A Look Back, A Look Ahead, 53 Pol. Res. Q. 62 (2000).
24 See generally, Supreme Court Decision-Making: New Institutionalist Approaches (Cornell W. Clayton & Howard Gilmann eds., Univ. Chicago Press 1999); Lee Epstein & Jack Knight, The Choices Justices Make (Congressional Quarterly Press 1998).
25 See Lee Epstein, Olga Shvetsova & Jack Knight, The Role of Constitutional Courts in the Establishment of Democratic Systems of Government, 35 Law & Soc'y Rev. 117 (2001). See also Jack Knight & Lee Epstein, On the Struggle for Judicial Supremacy, 30 Law & Soc'y Rev. 87 (1996) (applying an earlier version of this model to a study of the U.S. Supreme Court in the early 1800s).
Roux   |   Principle and pragmatism on the Constitutional Court of South Africa 113
In its application to constitutional courts in new democracies, this approach produces a theoretical model that counsels courts to ensure their decisions fall within the political branches' "tolerance interval" for every case that they decide.26 By so doing, the model holds, a constitutional court in a new democracy may build its institutional legitimacy to the point where it has wide discretion to decide cases in accordance with its policy preferences.27
Although the Epstein/Knight model is compatible with some of the more critical accounts of judicial review in legal theory,28 it presents two insuperable problems for liberal legal theory. First, it ignores, or at least downplays,29 the role of legal doctrine in narrowing the range of policy preferences that constitutional courts in new democracies may pursue. For liberal legal theorists, any attempt to understand the behavior of such courts must take into account the way legal doctrine constrains their capacity to act.30 Second, the Epstein/ Knight model dismisses out of hand the possibility that a constitutional court in a new democracy may have no concrete policy preferences in relation to a particular case.31 Of course, any decision by such a court expresses a policy preference in the weak sense that it distributes benefits and burdens among competing interest groups.32 It is important to most liberal legal theorists, however, that this fact should be seen as a mere side effect of politically "neutral"
26 Epstein et al., supra note 25, at 128-129. "Id. at 156.
28 See Duncan Kennedy, A Critique of Adjudication (Fin de Siecle) (Harvard Univ. Press 199 7). See also William N. Eskridge Jr. & Philip P. Frickey, The Supreme Court, 1993 Term Foreword: Law as Equilibrium, 108 Harv. L. Rev. 26 (1994) (applying the strategic approach in order to understand the U.S. Supreme Court's 1993 term).
29 See Epstein et al., supra note 2 5, at 12 9 (recognizing "case authoritativeness" as one of four elements impacting on the tolerance interval for a particular case, but otherwise ascribing no constraining role to law in the determination of judicial policy choice).
50 Of course, there is a range of views in liberal legal theory about exactly how constraining law is. Nevertheless, there is a crucial difference of opinion between liberal legal theory and Critical Legal Studies (CLS) over the minimum constraints that law must be assumed to impose for strong-form judicial review to be at all legitimate. See Kenn Kress, Legal Indeterminacy, 77 Calif. L. Rev. 283 (1989) (summarizing the CLS position on legal indeterminacy and defending law's claim to legitimacy in the face of such accounts). I locate my own attempted theorization of the problem posed in this essay within liberal legal theory, not because of any a priori commitment to that approach but because CLS's take on law's determinacy seems to me to preclude any interest in the way in which a constitutional court in a new democracy may go about establishing its legal legitimacy. If law is radically indeterminate, as CLS and most political science accounts assume, then the only issue of concern is how such a court establishes its institutional legitimacy.
51 I say "concrete" here because, as will become clear later, I do think that liberal legal theory is normatively compatible with certain types of long-range strategizing on the part of constitutional courts, such as the strategic pursuit of democratic consolidation.
52 Cf. Ronald Dworkin, A Matter of Principle 9 (Harvard Univ. Press 1985).
114 I-CON
January 2009
Vol. 7: 106
adjudication according to law.33 For liberal legal theory, the interesting question is not how constitutional courts in new democracies strategically assert their concrete policy preferences, but how such courts may remain faithful to the ideal of law-governed adjudication in sometimes very difficult institutional circumstances.
This is not a question that has received much attention in the legal literature,34 mainly because the focus of this literature has been on the legitimacy of judicial review in mature democracies, where a relatively high degree of institutional security may be assumed. As a starting point, however, the legal literature does suggest two theoretical possibilities, which may be thought of as occupying the opposite ends of a continuum.35 On one end of this continuum, Ronald Dworkin's theory of constructive interpretation provides, perhaps, the most forceful theorization of how a constitutional court, despite the inevitably political nature of its role, may remain faithful to the ideal of law-governed adjudication.36 On the other end of the continuum are those theories that are skeptical about the capacity of law to generate single right answers, but which are, nevertheless, liberal in the sense that they accept the value of the rule of law as one consideration among many that judges ought to take into account. Of these theories, the most instructive for my purposes is Richard Posner's restatement of the tenets of legal pragmatism, and, in particular, his (quite controversial) arguments about how appellate courts in the United States may legitimately take into account the systemic consequences of their decisions.37
53 See Herbert Wechsler, Toward Neutral Principles of Constitutional Law, 73 Harv. L. Rev. 1, IS (1959) (accepting that constitutional adjudication deals with political questions but arguing that ''what is crucial ... is not the nature of the question but the nature of the answer that may validly be given by the courts").
54 The exception is Ruti Teitel, Transitional Jurisprudence: The Role of Law in Political Transformation, 106 Yale L.J. 2009 (1997) (examining different conceptions of the rule of law in transitional democracies, though not providing a theory of judicial review as such). See also Ruti G. Teitel, Transitional Justice (Oxford Univ. Press 2000).
551 do not explore the implications of the claims of legal positivism for my research question in this essay, mainly because legal positivism lacks a detailed theory of constitutional adjudication. I also do not take seriously the possibility that, by deferring to the intentions of the constitutional drafters or, when the constitutional text is unclear, the preferences of the political branches, a constitutional court in a new democracy may build its legal legitimacy. The weaknesses of intentionalism are well known. See, e.g., Ronald Dworkin, Law's Empire 359-369 (Harvard Univ. Press 1986). The problem with the second, "passivist" approach is that it permanently relegates the court to an institutional position of no consequence. Id at 369-379.
56 See Dworkin, supra note 32, at 33. See also Wechsler, supra note 33, at 15-16.
57 See Richard Posner, Law, Pragmatism and Democracy (Harvard Univ. Press 2005). Posner's theory is most famously illustrated by his discussion of Bush v. Gore, 531 U.S. 98 (2000), which he sees not as a decision that bordered on the illegitimate by reason of its disrespect for settled precedent but as one laudable for its pragmatic pursuit of constitutional stability. Id. at 322-356.
Roux   |   Principle and pragmatism on the Constitutional Court of South Africa 115
Like Dworkin's theory, Posner's argument is not directed at constitutional courts in new democracies and, therefore, requires some extrapolation. To operate as a prescription for how such courts should behave one would have to graft onto Posner's theory the idea that the preservation of its institutional security might be something a constitutional court in a new democracy could strategically pursue. In every case that comes before it, so this revised theory would hold, a constitutional court in a new democracy should assess the likely impact of its decision on its institutional security and then decide the case in a way that optimally balanced its need for legal legitimacy with its ability to continue functioning.
In an intriguing passage, toward the end of Law's Empire, Dworkin appears to make some concession to the strategic nature of judicial review when he remarks that
[a]n actual justice must sometimes adjust what he believes to be right as a matter of principle, and therefore as a matter of law, in order to gain the votes of other justices and to make their joint decision sufficiently acceptable to the community so that it can continue to act in the spirit of a community of principle at the constitutional level.38
If one understands the highlighted passage as referring to something like the strategic objectives the revised Posnerian theory would advise judges to promote, then the difference between the two approaches narrows considerably. The precise import of Dworkin's concession is that principled adjudication is an ideal to which individual judges should aspire; however, it is one that may be compromised for two reasons: (1) to convince a sufficient number of the judge's colleagues to support a weaker version of the principle; and (2) to make the decision "more acceptable to the community." The thinking behind the first reason is evidently that a weaker principled decision by a majority of judges may, on occasion, be better, strategically speaking, than a stronger principled decision by a judicial minority. The second reason contemplates that a judge, in compromising on principle in order to win the support of fellow judges, may be motivated, additionally, by the thought that her community might not be capable of accepting the more strongly principled decision. Rather than risk rejection of her decision by that community, Dworkin suggests, a judge may prefer to "adjust" her reasons for a decision in order to ensure its acceptance by the community and, by this device, the continued functioning of her community as a community of principle.
Dworkin does not go on to explain what he means by the term "adjust" or the precise circumstances in which compromises on principle would be justified in order to keep the ideal of a community of principle alive. Self-evidently, any "adjustment" of a principle could not extend to its abandonment, since that would do violence to the term and make the last part of the passage, in
58 Dworkin, supra note 35, at 380-381, emphasis added.
116 I-CON
January 2009
Vol. 7: 106
which the community of principle is assumed to continue, contradict the first. More than this, however, we cannot say. In any case, Dworkin's argument, as noted earlier, is not directed at constitutional courts in new democracies. But what if it were? What sorts of adjustments and what sorts of circumstances could one envisage? Imagine that a case came before a court that was plainly controversial, either because the principled outcome contradicted the political branches' express policy preferences or because a significant section of the population was known to be opposed to it. In this situation, a court composed of judges who were individually committed to deciding cases on principle might find itself weighing the consequences of deciding this particular case on principle against the long-term institutional costs of such a decision. Possessed of certain knowledge that a case like this, if decided on principle, would bring the court into institution-threatening conflict with the political branches, some of the judges might decide to compromise on principle or to hand down a decision less forceful, as a matter of principle, than it otherwise might have been. If pressed, the judges who joined such a decision might justify their behavior as being in the overall interests of the constitutional project, arguing that the success of that project depended on their preparedness to make pragmatic compromises of this sort. "We decided the case this way," one might imagine them saying, "in order to survive to fight another day."
Of course, the decision whether or not to compromise on principle in a particular case would depend on a very difficult judgment call. Some highly charged cases would present questions of principle that could not be avoided, even if it were to mean the immediate closing down of the court or the replacement of its judges by more compliant ones. One could think here, for example, of a constitutional challenge to conduct on the part of the political branches that went to the very heart of the constitutional project. Compromising on principle in such a case would be self-defeating for a constitutional court, since it could never hope to recover its reputation for principled adjudication and, in any case, the constitutional project would have no point if it required a court to compromise on principle in such cases. Between this type of case, however, and the many routine cases that constitutional courts in new democracies decide, there would be other, less clear-cut cases that a court might think were not worth deciding on principle for fear of the consequences. Examples here might be a difficult case involving the conduct of foreign relations or the structure of the electoral system. In respect of these cases, the court might decide that a tactical retreat from principle was required, thus avoiding confrontation with the political branches and permitting the court to build its legal legitimacy through principled adjudication in other, less controversial cases.
In addition to this type of strategic behavior in controversial cases, one might imagine a constitutional court in a new democracy developing its jurisprudence in routine cases in such a way as to create greater discretion for itself. For example, a court might devise, in the first case that came to it under a particular section of the bill of rights, a context-sensitive review standard that
Roux   |   Principle and pragmatism on the Constitutional Court of South Africa 117
enhanced its ability to decide later cases on their particular facts. Similarly, a court, aware of the implications of an expansive, principled decision for later cases, might develop a collective judicial ethic of saying only as much as necessary to dispose of a case. Cass Sunstein has described this kind of strategy as akin to the making of "incompletely theorized agreements."39 According to this understanding, the output of a constitutional court in terms of principle is necessarily less than the sum of its parts since principled judges do not always agree, either among themselves or with competing principled views in their community. If true, the record of a constitutional court in a new democracy should resemble less the triumphant march of a forum of principle than the cautious output of a group of judges collectively sensitive to the need for a certain unanimity of purpose.
Section 3 of this essay seeks to illustrate that something like this mix of principle and pragmatism provides the best explanation for the CCSA's record to date. Three sets of cases are discussed. In the first set of cases, involving the constitutionality of the death penalty, the legalization of same-sex marriage, and the provision of antiretroviral treatment to pregnant women, the CCSA was able to hand down legally credible decisions in circumstances that were not obviously favorable to principled decision making. From the political science perspective, such decisions should not have been possible, since the CCSA had not, by the time of these decisions, built the institutional legitimacy required to assert its policy preferences in this way. In terms of the theoretical framework developed in this essay, however, this set of cases can be explained. In all three instances, the discussion shows, the CCSA was able to exploit the peculiar relationship between public support and institutional security in South Africa to hand down decisions of principle and, in this way, build its legal legitimacy.
In the second set of cases, the political context was less propitious, and the CCSA can be seen to have compromised on principle in order to avoid confrontation with the political branches. As a proportion of its total record, the number of these cases is very low. Nevertheless, this set of cases should not be dismissed as mistakes (rather than pragmatic compromises), for two reasons: first, because in two of the decisions a judge writing for the minority set out principled arguments that the majority was at liberty to accept but chose not to; and, second, because close analysis of these decisions reveals both an acute awareness on the part of the Court of the political stakes and an attempt by the Court to use technical legal devices to minimize the impact of these cases on its institutional security.
In the third and final set of cases, the CCSA's pragmatism manifests itself in the form of doctrinal choices aimed at maximizing its discretion to decide later
59 See Cass R. Sunstein, Legal Reasoning and Political Conflict (Oxford Univ. Press 1996). For an application of Sunstein's work to the record of the CCSA, see Iain Currie, Judicious Avoidance, 15 S. Afr. J. Hum. Rts. 138 (1999).
118 I-CON
January 2009
Vol. 7: 106
cases on their particular facts. In legal maneuvers that largely escape the attention of political science models, the CCS A can be seen to have chosen standards of review that will progressively reduce the tension between the need to establish its legal legitimacy and concerns about its institutional security.
3. Assessing the record of the CCSA
3.1. Principled decisions in the face of contrary public opinion and determined political opposition
The record of the CCSA contains a number of decisions that bear out the supposition, canvassed in section 1, that a constitutional court in a new democracy may be able to ignore public opinion as a limit on principle in certain circumstances. The case that best illustrates the point is the CCSA's decision on the constitutionality of the death penalty, State v. Makwanyane40 The use of the death penalty against political criminals and its disproportionate use with respect to black offenders in ordinary criminal matters was one of the most grievous of apartheid's many evils. There was, thus, a strong view prevailing among the ANC political elite at the time of the transition to democracy that the death penalty should be abolished.41 At the same time, however, South Africa's high rate of violent crime and generally conservative public attitudes on capital punishment meant that the vast majority of South Africans, including the ANC's political support base, favored the retention of the death penalty. In the result, the 1993 South African Constitution did not contain an express abolition clause along the lines of the German Basic Law.42 Instead, it included strong rights to life43 and to freedom from cruel, inhuman, or degrading punishment.44 The inclusion of these rights undoubtedly weighted the outcome in favor of abolition, but left the actual decision on the constitutionality of the death penalty to the CCSA to make.45
Although all of the judges in Makwanyane wrote separate concurring opinions, the opinion of Justice Arthur Chaslcalson, the president of the CCSA (as
40 1995 (3) SALR 391 (CC) (hereinafter Makwanyane). This was the first case heard by the CCSA and the second to be decided.
41 See Richard Spitz with Mathew Chaskalson, The Politics of Transition 331 (Witwatersrand Univ. Press 2000) (noting that "[t]he ANC's wish for capital punishment to be abolished was well known, and is clearly stated in its draft Bill of Rights").
42 Art. 102.
43 S. Afr. (Interim) Const., 1993, § 9.
44 Id. at § 11(2).
45 See Makwanyane, at Iff 20-25 (describing the 1993 Constitution's failure to deal with the issue of capital punishment as "not accidental" and locating the CCSA's power to decide the issue in the delegation of it to the Court to decide). See Heinz Klug, Striking Down Death, 12 S. Afr. J. Hum. Rts. 61, 65 (1996) (describing the delegation of so important an issue to the CCSA as "extraordinary" and reflective of "a failure to understand the delicate institutional role the Court will find itself in").
Roux   |   Principle and pragmatism on the Constitutional Court of South Africa 119
the position of chief justice was then known), is the most significant for present purposes, not just because it was the main judgment but also because it contained the leading discussion of the relevance of public opinion to the case.46 In argument, the attorney general had contended that statistical evidence showing strong public support for the death penalty in South Africa should have some bearing on the CCSA's interpretation of the constitutional rights at issue. Justice Chaskalson rejected this contention in strident terms:
Public opinion may have some relevance to the inquiry, but in itself, it is no substitute for the duty vested in the courts to interpret the Constitution and to uphold its provisions without fear or favour. If public opinion were to be decisive there would be no need for constitutional adjudication. The protection of rights could then be left to Parliament, which has a mandate from the public, and is answerable to the public for the way its mandate is exercised, but this would be a return to parliamentary sovereignty, and a retreat from the new legal order established by the 1993 Constitution.47
For South African lawyers, steeped in a formalist legal culture, this passage would not have been all that noteworthy as a statement of constitutional law doctrine.48 Of course, the CCSA, having been tasked with the duty of enforcing constitutional rights, should do so without regard to extrinsic, nonlegal considerations like public opinion. From the perspective of the CCSA's interest in the preservation of its institutional security, however, this statement is remarkable for the forcefulness of its rejection of the relevance of public opinion to the case. How could the CCSA afford to be so cavalier about the loss of public support that its decision to abolish the death penalty would surely trigger?49
46 See Makwanyane, at Iff 87-89. Other judgments in this case that largely followed Justice Chaskalson's lead on the relevance of public opinion include Justice Didcott's judgment (at f 188) and Acting Justice Kentridge's somewhat more qualified judgment (at Iff 200-201).
47 Makwanyane, at f 88.
48 For academic commentary on this passage, see Alfred Cockrell, Rainbow Jurisprudence, 12 S. Afr. J. Hum. Rts. 1, 18-19 (1996) (arguing that this passage embodies the CCSA's "official position" on the place of "popular morality" within substantive constitutional reasoning); Myron Zlotnick, The Death Penalty andPublic Opinion, 12 S. Afr. J. Hum. Rts. 70, 73 (1996) (noting that Justice Chaskalson does concede that public opinion might have "some relevance to the inquiry" but pointing out that he does not in the end indicate "what weight is to be given to public opinion"); Max du Plessis, Between Apology and Utopia: The Constitutional Court and Public Opinion, 18 S. Afr. J. Hum. Rts. 1, 2 (2002) (citing this passage in the context of an article explaining how the CCSA has sought to educate the public through the use of critical morality).
49 Cf. Klug, supra note 45, at 62 (arguing that the "blunt dismissal" of the relevance of public opinion in Justice Chaskalson's judgment was "mediated" in other judgments by "recognition of a national will to transcend the past"); Hugh Corder, Judicial Authority in a Changing South Africa, 24 Legal Stud. 253, 268 (2004) (arguing that the CCSA attempted to "trigger" public support for its decision elsewhere in the judgment by grounding its decision in "traditional African concepts of human solidarity").
120 I-CON
January 2009
Vol. 7: 106
The answer to this question, as suggested in section 1, lies in the fact that in a democracy like South Africa, with one dominant political party, a constitutional court may find that its institutional security is relatively immune to low public support. Where a single party dominates a country's electoral politics, lack of public support for a constitutional court is unlikely ever to be translated into votes for a rival political party. In such a situation, the court may be able to ignore public opinion as a limit on principle, provided that the dominant political party insulates it from the immediate repercussions of its decision.
This was precisely the context in which Makwanyane was decided. The negotiations leading up to the adoption of the 1993 Constitution had failed to produce a clear consensus on the abolition of the death penalty, at least in part because the ANC political elite was at odds with its supporters. In these circumstances, there was almost no cost to the CCSA in institutional terms in striking down the death penalty; the ANC elite favored this outcome and was content for the CCSA to take the burden of the decision onto itself. On the other hand, there was a considerable degree of legal legitimacy to be gained in writing a decision based on principle. Reading the passage quoted above, it is almost as if Justice Chaskalson deliberately uses the attorney general's argument as a foil to define the CCSA's institutional role, sending out a clear signal in this, the second case to be decided by the Court, that its claim to legitimacy would be based on strict adherence to the law/politics distinction.50 In a different political setting—one in which the CCSA's institutional security was dependent on public support—this strategy might have backfired. In the special circumstances of the Makwanyane case, however, the strategy arguably succeeded. Although the CCSA undoubtedly lost support as a result of this decision,51 its institutional security was never threatened, and its legal legitimacy was considerably enhanced.
At first glance, the other main issue of constitutional principle on which South African public opinion is overwhelmingly conservative—gay and lesbian rights—seems to share none of the same features as the death penalty. Both the 1993 and the 1996 constitutions mention sexual orientation as a ground of unfair discrimination.52 Thus, the constitutional status of gay and lesbian rights was not really left to the CCSA to decide; the battle of principle was mostly won by gay and lesbian groups in the constitutional negotiations
50 Cf. Karl E. Klare, Legal Culture and Transformative Constitutionalism, 14 S. Afr. J. Hum. Rts. 146, 173 (1998) (arguing that the rejection of public opinion in Makwanyane flows from the CCSA's desire to "buttress the legitimacy - the 'law-ness,' if you will - of its decision by repeated affirmation of the law/politics distinction").
31 See Du Plessis, supra note 48, at 5-6 (reporting on the results of a survey conducted in December 1995, six months after the decision in Makwanyane was handed down, showing 75 percent support for the reintroduction of the death penalty).
52 S. Afr. (Interim) Const. 1993, § 8(2); and S. Afr. Const.1996, § 9(3).
Roux   |   Principle and pragmatism on the Constitutional Court of South Africa 121
process,53 with litigation under both the postapartheid constitutions mainly about bringing old-order South African legislation into line with the constitutional compact.54
Nevertheless, the way in which this litigation progressed is pertinent. The key case of Fourie55 involved a constitutional challenge to the common law definition of marriage and to a provision of the Marriage Act that made the institution of marriage the exclusive preserve of heterosexual couples.56 Given the CCSA's past decisions, and the inclusion of sexual orientation as a ground of unfair discrimination in the 1996 Constitution, the question of substantive law was never in doubt.57 What was not a foregone conclusion, however, was the nature of the remedy the CCSA would devise. In an opinion written by Justice Albie Sachs,58 the majority of the Court reasoned that, since the issue of same-sex marriage was one of "status," the Court's remedy needed to be "secure," and that the best way of ensuring this outcome was to give Parliament an opportunity to amend the Marriage Act so as to provide for same-sex marriage.59 In lone dissent, Justice Kate O'Regan argued that the Court's duty to provide appropriate relief meant that any delay in the amendment of the marriage laws would amount to a failure of constitutional justice.60
The difference between the majority and minority judgments was not really a difference of substance, since Justices Sachs and O'Regan were agreed that the marriage laws needed to be amended.61 Rather, the difference between the two judgments concerned the appropriate institutional role of the CCSA in effecting this amendment. For Justice Sachs, the applicants' equality claims were best served by "respecting the separation of powers and giving Parliament an opportunity to deal appropriately with the matter."62 This was a clever conceit, implying, as it did, that constitutional rights may sometimes be better
53 See Spitz, supra note 41, at 306-307 (describing the history of the inclusion of "sexual orientation" as a listed ground of unfair discrimination in the 1993 Constitution).
54 See National Coalition for Gay andLesbian Equality v Minister of Justice 1999 (1) SALR 6 (CC) (challenge to various laws criminalizing sodomy); National Coalition for Gay and Lesbian Equality v Minister of Home Affairs 2000 (2) SALR 1 (CC) (challenge to immigration legislation).
55 2006 (1) SALR 524 (CC).
56 Marriage Act 25 of 1961 § 30.
57 For a full discussion of the legal issues in this case, see Beth Goldblatt, Same-sex Marriage in South Africa: The Constitutional Court's Judgement, 14 Feminist Leg. Stud. 261 (2006).
58 Fourie, at ff 115-161.
59 Id. atf 136.
50 Id. at 1170.
51 Justice O'Regan dissented only from the remedy ordered by the majority, not the reasoning in respect of the substantive issues at stake. Id. at Iff 163-165.
52 Id. atf 139.
122 I-CON
January 2009
Vol. 7: 106
vindicated by legislative amendment than by judicial fiat, and, therefore, that the separation-of-powers doctrine and the CCSA's duty to enforce constitutional rights are not always in conflict. Justice O'Regan's response to this point had two aspects, combining a "justice delayed is justice denied" argument with an argument that the judicial amendment of South Africa's marriage laws would not necessarily mean disrespect for the separation of powers.63 There would be nothing to prevent Parliament, Justice O'Regan noted, from later amending the Marriage Act according to its own sense of the constitutional rights at issue, provided that the amendment was in line with the principles laid down in Fourie.64
Although the formal difference between the two Fourie judgments thus had to do with the requirements of the separation-of-powers doctrine, it is clear that the legal rules on this point were fairly indeterminate. Neither of the judgments was obviously more persuasive than the other. Instead, what separates the judgments is a difference of opinion concerning the way in which the CCSA should go about building public support for decisions of constitutional principle. For Justice Sachs, it was important for the CCSA to enlist the legislature's cooperation in the enforcement of a legal change that was likely to be highly divisive, and that ran the risk of further weakening public support for the Court. For Justice O'Regan, the constitutional text was sufficiently clear to suggest that the CCSA would lose legal legitimacy if it did not grant the applicants an immediate remedy, and that this was, ultimately, a more important factor in securing public support for the Court.65
Whichever of the two views is correct, concern regarding the impact of the CCSA's decision on its public support seems to have figured more prominently in Fourie than it had in Makwanyane. And yet the ANC's electoral dominance, if anything, had become more entrenched in the ten years separating these decisions.66 How can this change in the judges' attitude be explained? Part of the answer, it is suggested, lies in a difference in the ANC political elite's capacity to shield the CCSA from the repercussions of its decision. Unlike the death penalty, the issue of gay and lesbian equality is one on which there is considerable disagreement within the ANC political elite.67 It was, therefore, important
53 Id. at 1170.
54 Id. at 1169.
55 Id. at 1f 171. See also Justice O'Regan's judgment at Iff 166-167 (recognizing "the important democratic and legitimating role of the Legislature in our society" but arguing that, since the definition of marriage had been developed by the courts at common law, the "responsibility" for changing the definition in line with the Constitution lay primarily with the courts).
56 In the first democratic election in 1994, the ANC's support stood at just over 62 percent. In the last general election, in 2004, the ANC won just short of 70 percent of the votes.
57 The ANC supported the inclusion of "sexual orientation" as a ground of unfair discrimination in the 1993 Constitution (see Spitz, supra note 41, at 306-307), even though some of its leading members, including the current party president Jacob Zuma, are openly homophobic.
Roux   |   Principle and pragmatism on the Constitutional Court of South Africa 123
for the majority to frame the Court's order in a way that would embed the decision in democratic politics and disassociate recognition of same-sex marriage from the CCSA as far as possible.
Several other decisions might be adduced to illustrate the CCSA's relative capacity to decide cases on principle in the face of countervailing public opinion;68 however, enough has been said to make the point. In concluding this section, it will be instructive to discuss a case in which the usual relationship between public support and the CCSA's institutional security was reversed. On this occasion, by the time the CCSA came to make its decision, public opinion had been mobilized overwhelmingly in support of the principled outcome, while the ANC political elite was, for once, isolated.
The facts surrounding the CCSA's decision in Treatment Action Campaign69 are notorious but bear repeating. The case concerned a constitutional challenge to the government's program for the prevention of mother-to-child transmission of HIV, the virus that causes AIDS. The program was based on the supply of a particular antiretroviral drug, nevirapine, which controversially had been made available only at a select number of research and training sites. The Treatment Action Campaign, a well-mobilized, politically astute, and charismatically run social movement, challenged the restriction on the drug's availability under sections 27 {\) (a) and (2) of the 1996 Constitution, which guarantee everyone "the right to have access to ... health care services," while limiting the state's obligation to fulfill this right in certain respects.70 In its earlier decision in Grootboom,71 the CCSA had decided that the standard of review in such cases was "reasonableness," but had restricted itself to an order declaring the challenged program unconstitutional, without mandating specific relief. In Treatment Action Campaign, the constitutional claimants applied for an order that the government's mother-to-child prevention program be extended to everyone. The intrusive nature of the remedy sought, together with a climate of public distrust over the ANC government's policies on AIDS, made Treatment Action Campaign one of the most politically controversial cases to come before the CCSA in the first ten years of its existence.
Although there is no social survey evidence for this view, most commentators accept that, at the beginning of the Treatment Action Campaign case, and certainly by its conclusion, there was overwhelming public support for the
58 See Azanian People's Organisation (AZAPO) v President of the Republic of South Africa 1996 (4) SALR 671 (CC) (constitutionality of the Promotion of National Unity and Reconciliation Act 34 of 1995, in so far as it extended immunity from civil and criminal prosecution to the perpetrators of apartheid human rights abuses); Christian Education South Africa v Minister of Education 2000 (4) SALR (CC) (constitutionality of corporal punishment in schools).
59 2002 (5)SA721 (CC).
70 Subsection (2) provides that: "The state must take reasonable legislative and other measures, within its available resources, to achieve the progressive realization" of this right.
71 2001 (1) SALR 46 (CC).
124 I-CON
January 2009
Vol. 7: 106
extension of the government's antiretroviral program.72 In the mainstream newspapers, President Thabo Mbeki's dissident views on AIDS had long been the subject of public ridicule, with the president variously caricatured as an obsessive, late-night internet surfer or as a Stalinist enforcer, silencing dissent in his party by the sheer power of his personality.
After an inconclusive exchange of legal letters, the applicants launched their case in the Pretoria High Court, where they won a wide-ranging order mandating the government to supply nevirapine forthwith or other suitable drugs if medically indicated.73 Despite the forcefulness of the High Court's reasoning, it was obvious to all that the real battle would be fought in the CCSA, where the government clearly anticipated getting a more sympathetic hearing. The state's case was duly founded on strongly worded separation-of-powers arguments, the emotive undercurrent of which was betrayed by a statement by the minister of health on national television that she would disobey the CCSA's decision if it went against her.74 Although the minister was almost immediately forced to retract this statement, the ANC political elite's opposition to the Treatment Action Campaign case could not have been clearer.
Against this politically fraught background, it is remarkable that the CCSA's eventual decision not only vindicated the constitutional rights at issue but did so with apparent ease. Dismissing the state's separation-of-powers arguments as irrelevant and ill founded,75 the CCSA handed down a unanimous decision, in the name of the Court as a whole, declaring the restriction of the government's mother-to-child prevention program to the research and training sites unconstitutional and a violation of the applicants' right to have access to health care services. The Court imposed a wide-ranging order, with immediate effect, mandating the provision of nevirapine in all public hospitals, where medically indicated.76
Like the U.S. Supreme Court's decision in Brown v. Board of Education,77 the CCSA's decision in Treatment Action Campaign must be explained by anyone who hopes to offer an adequate theoretical account of its record. From the perspective of this essay, the explanation for the apparent ease with which the CCSA dealt with the Treatment Action Campaign case lies in the fact that the decision was less politically awkward than at first appeared.
72 See, e.g., Mark Heywood, Preventing Mother-to-child HIV Transmission in South Africa: Background, Strategies and Outcomes of the Treatment Action Campaign's Case against the Minister of Health, 19 S. Afr. J. Hum. Rts. 278, 306 (2003).
73 Treatment Action Campaign v Minister of Health 2002 (4) BCLR 356 (T).
74 See Heywood, supra note 72, at 308.
75 See Treatment Action Campaign at Iff 96-106. 76Id. atf 135.
77 Brown v. Board of Education of Topeka, 347 U.S. 483 (1954).
Roux   |   Principle and pragmatism on the Constitutional Court of South Africa 125
As noted above, the applicants in Treatment Action Campaign had pursued an effective mass mobilization strategy, through which they had succeeded in creating a groundswell of public support for the principled outcome. In response to this strategy, key members of the ANC government had already begun to break ranks with the president and his health minister by the time the CCSA delivered its judgment.78 In particular, three months before the case was heard in the CCSA, the premier of Gauteng, South Africa's wealthiest province, had announced that his government would implement a comprehensive antiretro-viral program to prevent the mother-to-child transmission of HIV. Although the minister of health criticized this decision and forced the premier to give the appearance of backing down, the implementation of the program continued.79 By the time the CCSA came to hand down its decision, therefore, President Mbeki and his health minister were politically isolated and had, in the view of one closely involved commentator, already lost the battle in the "court of public opinion."80 Far from inhibiting the CCSA's ability to hand down a principled decision, this situation made it much easier for the Court to enforce the Constitution. With the ANC government sliding toward an embarrassing political defeat, the Court's decision could even be said to have rescued it by providing an "objective" legal basis for the reversal of its policies.
The CCSA's decision in Treatment Action Campaign thus may be read as a mirror image of its decision in Makwanyane. Whereas the latter decision shows how the CCSA was able to use the ANC political elite's opposition to capital punishment to shield it from adverse public opinion, Treatment Action Campaign is the best example of a case in which the CCSA relied on favorable public opinion to enforce the Constitution in the face of significant opposition by the ANC political elite. The common thread running through the two decisions is the CCSA's ability, on occasion, to exploit the political context to hand down decisions of constitutional principle and, in this way, to build its legal legitimacy. The next section discusses three controversial cases in which the political context was less propitious, and in which the CCSA was, therefore, forced to compromise on principle in order to avoid direct confrontation with the political branches.
3.2. Cases in which the CCSA compromised on principle
In its record to date, the CCSA has delivered three decisions that stand out from the rest as compromises on principle. Two of these decisions dealt with the structure of the electoral system, and the third with foreign policy. In other jurisdictions, such as the United States, the issues these cases presented might have been treated as political questions and, therefore, beyond the Court's
78 See Heywood, supra note 72, at 292 (commenting that the "disjuncture between the provinces was to be the undoing of the government's legal case").
79 Id. at 304. 50 Id. at 306.
126 I-CON
January 2009
Vol. 7: 106
remit.81 In South Africa, where the CCSA's constitutional mandate is very clear, the CCS A was not able to avoid deciding these cases.82 Instead, it resorted to a variety of technical legal devices, most notably its separation-of-powers doctrine, to minimize the impact of these cases on its institutional security.
The first such case, New National Party of South Africa v. Government of the Republic of South Africa,83 concerned an application by a minority political party (the successor party to the party that had ruled South Africa under apartheid) for an order declaring certain sections of the Electoral Act84 unconstitutional against the right to vote. The impugned sections, which had been enacted only nine months before the 1999 general election, provided that citizens who wanted to register as voters on the national common voters' roll and to vote in an election had to be in possession of a particular kind of identity document. Surveys conducted before the law came into effect revealed that five million otherwise eligible voters, constituting 20 percent of the voting population, did not have the requisite document.85 On its face, therefore, the applicant's case challenged an apparently race-neutral electoral rule that prevented a significant number of citizens from voting. Underlying the case, however, was the fact that many of the people who were prevented from voting were members of South Africa's white minority and, as such, were likely to vote for the applicant or one of the other minority parties.
In a somewhat convoluted majority opinion, Justice Zak Yacoob held that the requirements for registering as a voter and voting in the national elections should be understood as measures taken to facilitate the exercise of the right to vote, rather than as limitations on the right.86 The appropriate standard of review for the case, therefore, was whether there was a "rational relationship between the [electoral] scheme ... and the achievement of a legitimate governmental purpose."87 In addition to, or in specification of, this standard (it is not
51 See Louis Henkin, Is There a "Political Question" Doctrine?, 85 Yale L.J. 597 (1976); Mark Tush-net, Law and Prudence in the Law of Justiciability: The Transformation and Disappearance of the Political Question Doctrine, 80 N. Car. L. Rev. 1203 (2002).
52 The CCSA has never formally considered whether the political question doctrine could be applied in South Africa, but the thrust of the jurisdictional provisions in section 167 of the 1996 Constitution makes this possibility extremely unlikely. See L.W.H. Ackermann, Opening Remarks on the Conference Theme, in A Delicate Balance: The Place of the Judiciary in a Constitutional Democracy 10 (Jonathan Klaaren ed., Siberlnk 2006) (retired justice of the CCSA remarking that, "[i]n a substantive constitutional state such as ours, there can be no so-called 'political question' doctrine leading to a conclusion different to that dictated by the Constitution"). See also the cases discussed in note 104 infra.
53 1999 (3)SALR191 (CC).
54 Act 73 of 1998.
55 1999 (3) SALR 191 (CC), at ff 29-30. 86 id. at 115.
57 id. atf 19.
Roux   |   Principle and pragmatism on the Constitutional Court of South Africa 127
clear), Parliament, in designing the electoral scheme, was under a duty to ensure that "people who would otherwise be eligible to vote are able to do so if they want to vote and if they take reasonable steps in pursuit of the right to
vote."88
Unusually for the CCSA, the majority opinion did not cite a single authority in support of these two holdings. Moreover, apart from a few perfunctory remarks about the "importance of the right to vote,"89 the majority made no effort to develop a principled understanding of the right in its constitutional and political context. In fact, the only argument offered by the majority in support of its rational-basis standard of review was an attempt to rebut Justice O'Regan's argument, in dissent, that, given the importance of the right to vote, it was appropriate for the Court to set a stricter standard.90 The majority's response to this argument was to assert that the separation-of-powers doctrine placed an absolute bar on the power of courts to review statutory provisions on the grounds of reasonableness.91 This sweeping statement contradicts both the express language of the 1996 Constitution92 and the standard of review the Court later adopted in relation to socioeconomic rights.93 As Justice O'Regan pointed out,94 the centrality of the right to vote to the democratic process suggests that, if any right is deserving of stricter scrutiny, it should be this one.95
Reading Justice O'Regan's powerful dissenting opinion, it is difficult to come to any conclusion other than that the majority failed to give a principled reading of the Constitution. It is not just that the majority opted for a deferential standard of review; it is that the reasons in support of its preferred standard were so perfunctory. Having asserted the importance of the right to vote, the majority devised a review standard that allowed the state to condition the exercise of the right on obtaining an identity document the state was plainly not capable of providing to all potential voters in time for the next election. Not just
58 Id. at 121.
89 Id. at 111.
90 Id. at 124.
91 Id.
92 See S. Afr. Const. 1996 §§ 26(2) and 27(2) (expressly giving the Court the power to decide whether the state has adopted "reasonable legislative and other measures" progressively to realize socioeconomic rights). See also Justice O'Regan's opinion at 1 123 (pointing out that there are other rights, apart from the right to vote, "which contain broad equitable denning characteristics," and that there is, therefore, no hard-and-fast rule against the "inclusion of an equitable consideration at the threshold level of the right").
93 See Grootboom and Treatment Action Campaign.
94 New National Party 1999 (3) SALR 191 (CC) at 1122.
95 Cf. Jeremy Waldron, Law and Disagreement 282 (Oxford Univ. Press 1999) (calling the right to participate in the making of laws "the right of rights" and conceding the importance of the judicial protection of this right in a book otherwise skeptical of judicial review).
128 I-CON
January 2009
Vol. 7: 106
that, but the onus was placed on citizens to take "reasonable steps in pursuit of [their] right to vote."96 Constitutional rights, on this approach, are not really human rights at all, but entitlements that accrue to people because of what they do to deserve them.
It is impossible to prove, of course, that the majority's deferential approach in New National Party was influenced by its concern for the potential negative impact of the case on its institutional security; however, the circumstantial evidence is strong. Unlike the situation at the time of the Court's decision in Treatment Action Campaign, there was no groundswell of support for the principled outcome. Instead, the applicant was the successor party to the reviled National Party, which had ruled South Africa for forty-six oppressive years and taken it to the brink of civil war. The nature of the applicant's case was also deeply ironic, involving as it did the assertion of a right for which the applicant in its former political guise had shown little regard. Would the political branches accept that such an unsympathetic litigant was entitled to succeed, and, if not, would a decision to compromise on principle do more long-term damage to the constitutional project than could be offset by the institutional security gains to be made? On both counts, the Court in New National Party appears to have decided not.
The second politically controversial case in which the CCSA can be seen to have compromised on principle also concerned political rights—in this case, the right of voters to determine the conduct of their representatives between elections. The main applicant in United Democratic Movement v. President of the Republic of South Africa97 was, once again, a minority political party, this time a splinter party that had broken away from the ANC when its charismatic leader, a former homeland head of state, had been expelled. The case arose after the ANC, initially with the support of some of the other parties, tabled two constitutional amendments and two supporting statutes aimed at enabling members of the national, provincial, and municipal legislatures to cross the floor. As matters stood at that point, the 1996 South African Constitution stipulated that the electoral system should be based on proportional representation, with floor-crossing expressly banned in terms of an antidefection provision.98 A subclause in this provision, however, contemplated the amendment of the Constitution by ordinary legislation to allow members of the national and provincial legislatures to cross the floor, provided it was enacted "within a reasonable period after the new Constitution took effect." The second of the two supporting statutes at issue in United Democratic Movement was meant to bring about this amendment five years after the 1996 Constitution had come into effect. In a unanimous decision, the CCSA struck down the statute on the
96 1999 (3)SALR191 (CC) atf 21.
97 2003 (1) SALR495 (CC) (hereinafter United Democratic Movement).
98 Section 46(1), read with item 23A of Annexure A to Schedule 6 of the 1996 Constitution.
Roux   |   Principle and pragmatism on the Constitutional Court of South Africa 129
grounds that it had not been enacted "within a reasonable time";99 however, it left the other supporting statute and the constitutional amendments intact. Although the Court's decision thus went against the state in one respect, it erected no principled hurdle in the way of the amendment of the Constitution. As a result, floor-crossing became immediately permissible in the local government sphere and, after a further constitutional amendment to cure a defect in the first supporting statute,100 in the national and provincial legislatures as well.
The argument of principle raised by the applicants and rejected by the CCS A in United Democratic Movement was twofold: (1) whether a closed-list proportional representation system that allowed floor-crossing contravened the guarantee in section 1 (d) of the Constitution of a "multi-party system of democratic government"; and (2) whether it infringed the various political rights in section 19, including the freedom to make political choices.101 Writing unanimously as "the Court," the CCSA began its judgment by expressly addressing the public controversy surrounding the case:
This case is not about the merits or demerits of the provisions of the disputed legislation. That is a political question and is of no concern to this Court. What has to be decided is not whether the disputed provisions are appropriate or inappropriate, but whether they are constitutional or unconstitutional. It ought not to have been necessary to say this for it is true of all cases that come before this Court. We do so only because of some of the submissions made to us in argument, and the tenor of the public debate concerning the case which has taken place both before and since the hearing of the matter.102
The first striking thing about this passage is the Court's use of the term "political question," which it deploys in a nontechnical sense to mean the wisdom of a policy choice, irrespective of its justiciable content. "Nontechnical" because there is nothing in this judgment to suggest that this passage marked the introduction of a formal political question doctrine.103 Rather, what the CCSA appears to be doing, rhetorically, is denying, as it did in Makwanyane, the political nature
99 The Loss or Retention of Membership of National and Provincial Legislatures Act 22 of 2002.
100 Constitution of the Republic of South Africa Amendment Act 2 of 2003.
101 S. Afr. Const. 1996 §§ 74(1) and (2) for special majorities for constitutional amendments that impact on the founding values and constitutional rights.
102 United Democratic Movement at f 11.
103 In Ferreira vLevin N.O.; VryenhoekvPowellN.O. 1996 (1) SALR 984 (CC) at f 180, the majority held that: "Whether or not there should be regulation and redistribution is essentially a political question which falls within the domain of the Legislature and not the Court." The use of the phrase "political question" in this passage is similar to its use in United Democratic Movement, supra note 98, and should be read as meaning nothing more than that the decision whether or not to enact a particular statute is the prerogative of the legislature.
130 I-CON
January 2009
Vol. 7: 106
of its role. Clearly conscious of the controversy surrounding the case, the Court here proclaims that its institutional role is to keep law separate from politics, and to decide the case according to neutral principles.
At face value, then, this passage appears to confirm the CCSA's commitment to building its legitimacy through law-governed adjudication. But this is merely the surface appearance of things. Read in the context of the judgment as a whole, the underlying purpose of the passage is to equate the assessment of the merits of the disputed legislation with the realm of politics and, in this way, to justify a retreat from principle. By playing on the ambiguity of the word "merits," the Court implies that it lacks the institutional competence to decide the case. This is plainly not true. Under a supreme-law constitution, a court is constantly called on to assess the merits of legislation according to the standards embodied in fundamental rights and other constitutional norms.
That this passage is an early signal of the CCSA's intention to compromise on principle is confirmed by its eventual treatment of the substantive issues for decision. On the question whether the disputed legislation infringed the guarantee of a multiparty system of democratic government, the CCSA held that the term "democracy" was too indeterminate to permit it to substitute its view of the appropriate form of South Africa's electoral system for that of the legislature;104 and, on the question whether the disputed legislation infringed political rights, the CCSA held that such rights were relevant only at the time of elections, and that citizens accordingly had no right to control the conduct of their representatives once elected.105 Neither of these holdings finds very much support in the constitutional text,106 and both have since been contradicted by other decisions that articulate a deep, participatory conception of democracy more in keeping with South Africa's political tradition.107 Once again, therefore, the CCSA appears to have compromised on principle in a case in which (1) the risk to its institutional security was high; and (2) the principle at stake was not so vital that its denial would permanently set back the constitutional project.
The third case in which this strategy is apparent concerned the executive's conduct of foreign relations, the classic situation in which the political question doctrine has been applied in the United States. The applicants in Kaunda v.
104 UnitedDemocratic Movement at Iff 23-75.
105 Id. at 149.
106 See Theunis Roux, Democracy, in Constitutional Law of South Africa, supra note 3, at 10-65-10-68; GlendaFick, Elections, in Constitutional Law of South Africa, supra note 3 , at 29-17 (expressing concern that the CCSA does not indicate which aspect of democracy it is referring to).
107 See Doctors for Life International v Speaker of the National Assembly 2006(6) S ALR 416 (CC) (striking down several health-related bills for failure by the legislature to facilitate public involvement in its processes); Matatiele Municipality v President of the Republic of South Africa (2) 2007 (1) BCLR 47 (CC) (striking down a constitutional amendment for the same reason).
Roux   |   Principle and pragmatism on the Constitutional Court of South Africa 131
President of the Republic of South Africa108 had been arrested in Zimbabwe on suspicion of participating as hired mercenaries in a planned military coup against the president of Equatorial Guinea. Amid allegations of poor treatment in Zimbabwe and threatened extradition to Equatorial Guinea, the applicants launched an urgent application in the Pretoria High Court demanding that the South African government seek their release or extradition to South Africa. After the High Court dismissed the application, the applicants appealed directly to the CCSA for relief.
The issue of principle in Kaunda was whether the South African government was under a duty to intervene to safeguard the applicants from threatened violation of their human rights, either under the right to diplomatic protection at international law, or under municipal constitutional law. Dismissing the international law argument on the ground that the right to diplomatic protection was not an individually enforceable human right but a prerogative that could be exercised by a state at its discretion,109 the majority turned to what it said was the key question for determination in the case: the extraterritorial application of the Bill of Rights. Section 7(1) of the South Africa Constitution provides: "The Bill of Rights is a cornerstone of democracy in South Africa. It enshrines the rights of all people in our country and affirms the democratic values of human dignity, equality and freedom." Over another strong dissent from Justice O'Regan,110 the majority held that this provision should be interpreted literally to mean that South Africans enjoy the protection of the Bill of Rights only when they are physically in South Africa. The majority presented this reading as though no other interpretation of section 7(1) was even remotely possible: "The bearers of the rights are people in South Africa. Nothing suggests that it is to have general application, beyond our borders."111
In these two sentences, as with the New National Party decision, the majority's reliance on assertion rather than reasoned argument betrays its equivocation on the point of principle. In decision after decision, both before and after Kaunda, the CCSA has expressed a preference for interpreting the Bill of Rights and other constitutional provisions in a "generous" and "purposive" way.112 According to this approach, any ambiguity in the constitutional text must be resolved in favor of the interpretation that best gives effect to the purposes and values underlying the new constitutional order. And yet, in Kaunda, the majority relied on the literalist approach it elsewhere condemns. The ordinary
108 2005 (4) SALR235 (CC).
109 Id. at ff 23-29.
110 Id. at llf 212-271.
111 Id. at137.
112 See, e.g., State vZuma 1995 (2) SALR 642 (CC) at If 13-18; Makwanyane, at If 9-10; Executive Council, Western Cape Legislature v President of the Republic of South Africa 1995 (4) SALR 877 (CC) at 1100; Mashavha v President of the Republic of South Africa 2005 (2) SALR 476 (CC) at f 32.
132 I-CON
January 2009
Vol. 7: 106
meaning of the phrase "all people in our country" in the context in which it is used is not "all people physically present within South Africa" but "all South Africans, without exception." The stress in the phrase, in other words, falls on the word "all," not "in." At the very least, the phrase is ambiguous and thus needed to be construed purposively in the light of the Constitution's underlying values.
The majority's disinclination to undertake such a reading is indicative of its sensitivity to the separation-of-powers issues raised by the case.113 Denied the luxury of a political question doctrine, the majority used its literalist reading of section 7(1) to avoid the institutionally awkward consequences of the application of the Bill of Rights. Had the majority interpreted section 7(1) purposively to mean that the Bill of Rights binds the state in all its dealings with its citizens, wherever they happen to find themselves, it would have had little option but to enforce the state's obligation to provide diplomatic protection.
For Justice O'Regan, this conclusion followed not from a purposive reading of section 7(1), but from the insight that, in fact, the case had nothing to do with the extraterritorial application of the Bill of Rights.114 Since the officials responsible for applying the Bill of Rights were at all relevant times in South Africa, the real question was whether section 3 of the Constitution, which confers on all citizens "the rights, privileges and benefits of citizenship," imposed a duty on the state to provide diplomatic protection in circumstances where it was entitled under international law to do so.115 Reading the entire Constitution purposively, Justice O'Regan answered this question in the affirmative. When presented with clear evidence of a threatened breach of its citizens' human rights under international law, she held, the state was under a constitutional obligation to act.116 The only real question was whether it was competent for the Court, in light of the separation-of-powers doctrine, to enforce this obligation. Although primarily the responsibility of the executive, the conduct of foreign relations involved the exercise of public power and was thus subject to the Constitution.117 At best, from the state's point of view, the Court's lack of expertise meant that it should not lightly interfere with decisions made by the executive in this area. It did not follow, however, that the state's obligation to provide diplomatic protection was not justiciable.118
As was the case in New National Party and Fourie, therefore, the difference between the majority opinion in Kaunda and Justice O'Regan's dissenting opinion can be traced to a disagreement about how best to balance the competing
113 The majority acknowledges these concerns at If 19 of its judgment.
114 Kaunda 2005 (4) SALR 235 (CC) at f 231.
115 H. at 1237.
116 Id. at ff237-238.
117 Id. at 11243-244.
118 Id. at 1247.
Roux   |   Principle and pragmatism on the Constitutional Court of South Africa 133
concerns of legal legitimacy and institutional security. In all three of these cases, Justice O'Regan chose principle over pragmatism, refusing in New National Party to adopt the weaker standard of review that the majority deemed more appropriate to the highly charged nature of the case; refusing in Fourie to make any concession to the strategic need to embed the Court's remedy in democratic politics; and refusing in Kaunda to defer to the executive's prerogative in the conduct of foreign relations. If one were to examine these decisions from the perspective of the small-group dynamics on the Court, one might conclude that Justice O'Regan's freedom to decide these cases on principle was made possible by the majority's pragmatism, and that her approach might well have been more cautious had she not been writing in dissent. From the theoretical perspective adopted here, Justice O'Regan's repeated dissents may be attributed to her different conception of the CCSA's separation-of-powers doctrine. Whereas the majority in New National Party and Kaunda took the view that this doctrine supplied a legally valid reason for reducing the level of principle in those cases, Justice O'Regan's dissents are premised on a more absolutist conception. For her, where the issue for decision falls squarely within the Court's competence, as any issue involving the interpretation of the Bill of Rights must, the separation-of-powers doctrine has little relevance. At most, it requires the Court to be conscious of the possible impact of its decision on the political branches' ability to perform their constitutional functions. The doctrine can never be used, however, as a justification for compromising on principle.
3.3. The CCSA's preference for context-sensitive, multifactor balancing tests
The six cases considered thus far have all been cases of high constitutional moment, either because the principled decision ran counter to strongly held public attitudes, or because the principled decision threatened to bring the Court into direct confrontation with the political branches. In the absence of a political question doctrine, the CCSA was not able to avoid deciding these cases but had to work with the political context and the legal materials to ensure that the decision it took did not impact negatively on its institutional security.
In less controversial cases, one might expect that the CCSA would have been able to cast off these strategic concerns to hand down decisions of principle. The CCSA's jurisdiction, however, is restricted to constitutional matters.119 Even in routine cases, therefore, the CCSA declares constitutional law that it might be required, in later, perhaps more politically awkward cases, to follow. Given this background, it is, perhaps, not surprising that the CCSA's record in routine cases is not that different from its record in politically controversial cases. What differences there are have to do with the kinds of strategy that the CCSA has pursued. Instead of a case-by-case movement between principle and pragmatism, what we see in these cases is a general tendency to convert con-
119 SeeS. Afr. Const., 1996, § 167.
134 I-CON
January 2009
Vol. 7: 106
ceptual distinctions into multifactor balancing tests. In this way, the CCSA has been able, progressively, to reduce the tension between decisions of principle and the need to preserve its institutional security.
The best example of this strategy is First National Bank of SA Ltd t/a Wesbank v. Commissioner, South African Revenue Service.120 Although this case involved a challenge under the constitutional property clause, the type of challenge that certainly has the potential to be controversial in South Africa, First National Bank was the sort of case that might have arisen in any liberal democracy. The statutory provision impugned provided that the commissioner for the South African Revenue Service could attach and sell in execution property found in the possession or under the control of a customs debtor, including property belonging to a third person. The applicant, one of South Africa's largest financial institutions, brought its case after several of its motor vehicles, which it had leased to a customs debtor, were attached under this provision.
The first part of the property clause, section 25 of the 1996 Constitution, provides:
(1) No one may be deprived of property except in terms of law of general application, and no law may permit arbitrary deprivation of property.
(2) Property may be expropriated only in terms of law of general application—
(a) for public purposes or in the public interest; and
(b) subject to compensation, the amount, timing and manner of which must be agreed, or decided or approved by a court.
On their face, these provisions invited the CCSA, in this, the first property rights case decided under the 1996 Constitution,121 to build its jurisprudence around four conceptual distinctions:122 a distinction between interests that constitute property and those that do not; a distinction between the deprivation and expropriation of property; a distinction between arbitrary and nonar-bitrary deprivations of property; and a distinction between expropriations undertaken "for a public purpose or in the public interest" and those that are not. In the United States, conceptual distinctions like these have preoccupied the Supreme Court for some time, leading on occasion to arbitrary results.123 In
120 2002 (4)SALR 768 (CC).
121 The CCSA had earlier decided one property rights case under the 1993 Constitution, Harksen v Lane N.O. 1998 (1) SALR 300 (CC) (also a routine case, involving a challenge to a provision of the Insolvency Act 24 of 193 6 and decided mainly under the equality clause).
122 See Theunis Roux, The 'Arbitrary Deprivation' Vortex: Constitutional Property Law after FNB, in Constitutional Conversations (Stuart Woolman & Michael Bishop eds., Pretoria Univ. Law Press, forthcoming 2008).
123 The literature on this topic is vast. For a representative example, see Bruce Ackerman, Private Property and the Constitution (Yale Univ. Press 1977).
Roux   |   Principle and pragmatism on the Constitutional Court of South Africa 135
a decision that says much about its general approach to constitutional adjudication, the CCSA in First National Bank cut through all of these distinctions and reduced the constitutional property clause inquiry to a single, multifactor balancing test. Holding that expropriations were a form of deprivation and, therefore, that all challenges under section 25 had to be heard first under section 25(1),124 the Court remarked that the main issue to be decided in any challenge under the constitutional property case was whether the impugned law provided "sufficient reason" for the deprivation.125 This question, the Court continued, had to be decided by examining "a complexity of relationships," including "the relationship between means employed, namely the deprivation in question and ends sought to be achieved, namely the purpose of the law in question"; "the relationship between the purpose of the deprivation and the person whose property is affected"; and "the relationship between the purpose of the deprivation and the nature of the property as well as the extent of the deprivation in respect of such property."126
The last part of the test was even more open-ended, with the Court holding that the level of review in any particular case could fall anywhere between rationality and proportionality according to the nature of the property in question and the incidents of ownership affected. "Whether there is sufficient reason to warrant the deprivation," the Court concluded, "is a matter to be decided on all the relevant facts of each particular case."127 It is hard to conceive of a more flexible review standard than this one. In place of the conceptual distinctions that section 2 5 invited the Court to make, this test substitutes a discretionary standard that makes the outcome of future caseshighly unpredictable.128 Although the test, no doubt, will become more certain with time, there are enough variable elements in it to allow the Court to adjust the level of review and, therefore, its decision to virtually any contingency.
A similar strategy may be discerned in the CCSA's socioeconomic rights jurisprudence, though none of these decisions could be described as routine. In both Grootboom and Treatment Action Campaign, the applicants had argued that the CCSA should define the "minimum core content" of the rights at issue.129 As it did in First National Bank, the CCSA declined to adopt the conceptual test that conceding this argument would have required, substituting instead a
124 First National Bank, 2002 (4) SALR 768 (CC) at f 5 7.
125 Id. atf 100.
126 Id.
127 Id.
128 The unpredictability of the First National Bank test for arbitrariness is illustrated by the fact that, in the next constitutional property rights case to come before the CCSA, the two lower courts reached diametrically opposed results by applying different levels of review to roughly the same set of facts. See Mkontwana v Nelson Mandela Metropolitan Municipality 2005 (2) BCLR 150 (CC).
129 Grootboom at Iff 31-33; Treatment Action Campaign at Iff 26-39.
136 I-CON
January 2009
Vol. 7: 106
single, overarching review standard for reasonableness.130 Like its test for arbitrary deprivation of property, the CCSA's reasonableness standard in socioeconomic rights cases is extremely context-sensitive. In Grootboom, for example, the CCSA held that, when determining the reasonableness of state measures adopted to fulfill socioeconomic rights, the Court should have regard for the comprehensiveness and coherence of the program. The first issue was to be determined by reference to the geographic coverage of the program131 and the extent of involvement in it of different levels of government.132 The second issue depended on the potential of the program to realize the right in question, though it need not be the only or even the best means of doing so.133 Both of these factors clearly give the Court tremendous discretion to take the specific details of the challenged program into account.
Many South African legal academics were quite concerned, at the time, about the CCSA's rejection of the minimum core content approach, which they argued amounted to an abdication of its responsibility to enforce socioeconomic rights.134 As soon as one accepts, however, that the CCSA's strategy in these cases may have been to devise a review standard that allowed it greater flexibility to manage its relationship with the political branches, much of the force of this criticism falls away.135
4. Conclusion
The foregoing analysis of the strategies used by the CCSA to build its legal legitimacy while preserving its institutional security is not capable of demonstrating that the Court is in a better position today than it would have been in had it not engaged in such strategies. That conclusion depends on a counterfactual and must be left to speculation. However, insofar as the CCSA is today still handing down decisions that the political branches do not always like, the cases discussed in section 3 provide strong evidence that a willingness to
130 Compare Martin Shapiro's argument that all constitutional rights, in the end, reduce to reasonableness, in Martin Shapiro & Alec Stone Sweet, On LM, Politics and Judicialization 179 (Oxford Oniv. Press 2002).
131 Grootboom at f 3 7.
132 Id. at 140.
133 Id. at!41.
134 See David Bilchitz, Giving Socio-economic Rights Teeth: The Minimum Core and its Importance, 119
5. Afr. L.J. 484 (2002); David Bilchitz, Towards a Reasonable Approach to the Minimum Core: Laying the Foundations for Future Socio-economic Rights Jurisprudence, 19 S. Afr. J. Hum. Rts. 1 (2003); Danie Brand, The Proceduralisation of South African Socio-economic Rights Jurisprudence, or "What Are Socio-economic Rights For?," in Rights and Democracy in a Transformative Constitution (Henk Botha et al. eds., SUN Press 2003).
135 See Mark S. Kende, The South African Constitutional Court's Construction of Socio-economic Rights: AResponse toCritics, 19Conn. J.Int'lL. 617, 622 (2004).
Roux   |   Principle and pragmatism on the Constitutional Court of South Africa 137
compromise on principle in certain cases might offer a viable way for some constitutional courts in new democracies to safeguard their institutional security. More importantly, the South African case shows that such compromises do not necessarily damage a court's overall reputation for law-governed adjudication.
The key to the CCSA's success has been its ability to exploit the political context to hand down decisions of principle in cases where other courts might have balked. This is most evident in the Makwanyane and Treatment Action Campaign decisions, in which the CCSA first used the ANC's electoral dominance to hand down a principled decision that flew in the face of public opinion, and then used favorable public opinion to support a principled decision on the extension of the government's antiretroviral program. Without these decisions of principle in controversial cases, the CCSA would not have built the reputation for legally credible decision making it currently enjoys.136
In New National Party, United Democratic Movement, and Kaunda, where the political circumstances surrounding the CCSA's decision were less propitious, the Court can be seen to have compromised on principle. For the majority judges, these decisions were justified by the Court's separation-of-powers doctrine, which provided a legally valid reason for the strategic compromises the majority thought needed to be made. Justice O'Regan's dissents in New National Party and Kaunda, together with her dissent in Fourie, cast serious doubt, however, on the determinacy of this doctrine. As much as the Court might insist that its separation-of-powers doctrine is reducible to clearly defined rules, there are at least two versions of the doctrine in its case law. The first, which can be found in Justice O'Regan's dissenting judgments but also in the Court's unanimous judgments in Makwanyane and Treatment Action Campaign, focuses on the nature of the question to be answered. Where that question is one of rights interpretation, the fact that the constitutionally required decision may intrude into areas primarily reserved for the political branches is simply an inevitable side effect of the fulfillment by the Court of its constitutional mandate and, as such, may not be factored into the decision-making process. The second version of the doctrine, which was applied by the majority in New National Party and Kaunda, and by the unanimous Court in United Democratic Movement, holds that separation-of-powers concerns may be taken into account by the Court when deciding on the level of review to be applied, and that such considerations may legitimately trump arguments of principle relating to the importance of the rights at stake and their place in the constitutional normative order.
Given this level of indeterminacy, the explanation for the CCSA's record must lie outside formal legal doctrine (although the constraints imposed by
136 See Mark S. Kende, The Fifth Anniversary of the South African Constitutional Court: In Defense of Judicial Pragmatism, 2 6 Vt. L. Rev. 753, 766 (2002) (referring to the CCSA's decisions on the death penalty and gay and lesbian equality and arguing that "the Court has been pragmatic in selecting only a few cases on which to expend its institutional capital").
138 I-CON
January 2009
Vol. 7: 106
formal legal doctrine still form part of the explanation). As this essay has attempted to show, one explanation for the CCSA's record lies in the peculiar South African configuration of three factors: legal legitimacy, public support, and institutional security. Unlike other constitutional courts in new democracies, the CCSA has not needed to court public opinion in order to safeguard its institutional security. The main reason for this is that the ANC political elite has shielded the Court from the political repercussions of its most unpopular decisions, allowing it to build its legal legitimacy through principled decision making in several important cases. In return, the CCSA has been careful to manage its relationship with the political branches, retreating from principle where such compromises were in the long-term interests of the constitutional project. In this way, a mutually beneficial relationship has developed between the CCSA and the ANC government, with the CCSA's reputation for legally credible decision making lending considerable legitimacy to the ANC's social transformation policies, and the ANC government's continued respect for, and obedience to, the CCSA's decisions helping to cement the CCSA's reputation as one of the most successful of the post-1990 constitutional courts.